Mia-AiLab / GLM-5.3-Flash-EXL3-3.5bpw

Vision78B401Updated Oct 30, 2026

Downloads

40

Likes

1

Last updated

Oct 30, 2026

License

apache-2.0

Mia's AI Lab's own EXL3 quantization of zai-org/GLM-5.3-Flash. A smaller sibling of GLM-5.3-Flash-EXL3-4bpw-TensorFold: same format, same layout, same unquantized tensors. Only the routed experts are narrower: every down projection stays at 4 bits, and the gate and up projecti…

Tasks & library

  • image text to text

Tags

  • glm5_next
  • glm
  • exl3
  • quantized
  • moe
  • image-text-to-text

Technical specifications

Parameters78B
LibraryNot available
LanguagesNot available
Licenseapache-2.0